Demo abstract: towards extracting semantics by visualizing large traceroute datasets

Abstract

Many Internet mapping projects have used traceroute as a measurement primitive and have generated hundreds of millions of traceroute samples used, among many purposes, for network debugging, troubleshooting and Internet mapping. In this work, we advocate the use of visualization as a means to extract semantics from large sets of traceroute data collected from large-scale traceroute campaigns. Using subsets of these datasets consisting of all traceroutes traversing a particular Internet eXchange Point (IXP), we illustrate how visualization allows us to (1) examine existing large datasets in a different way and (2) present different views of the data that can provide new insights and lead to new questions.